JPH03233336A - Pressure transmitter - Google Patents

Abstract

PURPOSE:To improve working efficiency and to reduce a cost by finding the factor of a correction curve by supplying calibration input to specific arithmetic expression, specifying and calculating the correction curve which corrects the characteristic of a differential capacity sensor by using the factor. CONSTITUTION:Calibration pressure Pc is applied to the differential capacity sensor 29 for three times, and electrostatic capacitance C1, C2 are read in with the sensor 29 under the control of a CPU 37, and they are converted into digital signals Vc, and are stored in a RAM 35. An arithmetic operation is executed by an arithmetic program stored in a ROM 36 by using three pairs of data stored in the RAM 35 in such way, and the proper factors (a), (b), and (c) of the sensor 29 are defined, and they are stored in a prescribed area of the RAM 35. Secondly, correction arithmetic expression I is defined by using defined factors (a), (b), and (c). Correction arithmetic expression II stored in the RAM 35 is stored fixedly in the ROM 36 functioning as an EEPROM. After that, the CPU 37 calculates applied measuring pressure PM from a capacity signal VM obtained via an A/D converter 33 when the measuring pressure PM is applied, and outputs it via a D/A converter 44. Hereinafter, the pressure is measured repeatedly.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to a pressure transmitter that converts pressure (including differential pressure) into an electrical signal using a differential capacitance type sensor, and particularly relates to a pressure transmitter that converts pressure (including differential pressure) into an electrical signal using a differential capacitance type sensor. Improved pressure transmitter to compensate for and improve accuracy.

<Conventional technology> Generally, when converting pressure into an electrical signal using a capacitance type sensor, its input/output characteristics have nonlinearity, and each sensor has unique characteristics. .

Therefore, by applying a calibration pressure in advance, this input/output characteristic is stored in the memory of a microcomputer unit as data unique to each sensor, and output calculations are performed by the microprocessor unit using this unique data. . Therefore, in order to achieve accurate pressure/electrical signal conversion, this input/output characteristic must be accurately set in memory for each sensor.

Conventionally, in order to approximate the actual input/output characteristics of each sensor, setting is generally performed using a polygonal line approximation as shown in FIG. 6, for example.

In FIG. 6(a), the horizontal axis is the input calibration pressure, the vertical axis is the output electrical signal, the broken line shows the actual characteristics, and the dotted line shows the polygonal line approximation. When the polygonal line approximation is performed in this way, an additional error will be generated as an output error as shown in FIG. 6(b).

In addition, to set other input/output characteristics, the actual input/output characteristics of each sensor are approximated by a polynomial as shown in FIG. 7(a).

In FIG. 7(A), the horizontal axis is the input calibration pressure, the vertical axis is the output electrical signal, the broken line shows the actual characteristics, and the dotted line shows the polynomial approximation. Even when polynomial approximation is performed in this way, an additional error is generated as an output error, as shown in FIG. 7(b).

<Problems to be Solved by the Invention> However, when setting the input/output characteristics in the memory of the microcomputer unit in this way, an approximation formula is used, so it contains a considerable amount of error. In addition, in order to improve accuracy, in the case of polynomial approximation, it is necessary to shorten the straight line section and take fine bending points, so there is the inconvenience of having to increase the number of calibration points, and in the case of polynomial approximation, Since it is necessary to increase the degree of the polynomial, there is a problem in that the number of unknowns increases and the number of calibration points must be increased, resulting in a decrease in work efficiency.

Means for Solving the Problems> In order to solve the above problems, the present invention provides a method in which a first electrode and a second electrode are provided facing the moving electrode, and a sealing liquid is filled between them. a differential capacitance sensor forming first and second capacitances that differentially change according to the pressure to be detected; and converting capacitance signals corresponding to the first and second capacitances into digital signals. In the microcomputer unit that performs calculation processing, and in the pressure transmitter that converts the calculation results of this microcomputer unit into an analog signal and outputs it, the microcomputer unit performs known calibration at any three points within the pressure measurement range. The three capacitance signals Vc obtained for the pressure Pc are digitally converted and stored in a memory, and these stored data pairs (Pc, Vc
), the coefficients a, b of the formula Vc = [aPc/(b Pc')] + c derived from the principle of differential capacitance type
, c, and the coefficients a, b= determined in this way by the microcomputer unit.
Measurement in which the input/output characteristics are corrected using the formula % formula %)] (however, when VM: C, P bar = 0) from the capacitance signal VM obtained corresponding to the unknown measured pressure PM using c. The apparatus is also provided with a calculation means for correcting the input/output characteristics based on the pressure P.

Function> The known calibration pressure P at any three points within the pressure measurement range is determined by the coefficient setting means of the microcomputer unit.
For example, the three capacitance signals VC obtained for c are digitally converted and stored in a memory, and the stored data pairs (Pc, Vc) are used to calculate the coefficients a, b, c of the formula %. Determine.

Thereafter, the formula Pr+=[a± I (a2+4 ( V.M.C.
) 'b) 112]/-2(Va-C> (However, when VM=C, PM=O) Outputs the measured pressure P bar whose input/output characteristics have been corrected by calculation.

As described above, it is possible to obtain a highly accurate output signal for the measured pressure with a small number of calibration points.

Practical Examples> Examples of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of one embodiment of the present invention.

29 is a differential capacitance sensor, which is arranged opposite to the movable electrode 7 by a fixed amount &S, and has capacitances C4 and C, respectively.
2 is formed.

30 is a microprocessor unit. 31 is a capacitance/time converter that converts the capacitance i CI , C2 into a corresponding time signal, and 32 is a timer counter that converts the time signal into a digital value. Analog/
A digital converter 33 is configured. 35 is RAM (
Random access memory), 36 is ROM (read only memory), EEPROM (electronic errorable program ROM), or a combination of these, but it is typically handled as ROM, and the addressing of these is from the CPU (processor) 37. This is done via a bus 38 and a latch decoder 39. 40 is a data bus. The output data from the timer counter 32 is transferred to the RAM.
35.

A predetermined calculation program and initial data are stored in the ROM 36, and the ROM 36 is stored under the control of the CPU 36.
The result calculated according to the calculation procedure stored in R
It is stored in AM35. Note that the illustration of the control bus is omitted. The calculation result of the closest connection is converted into a duty signal by a timer/counter 41, and the duty signal is converted into an analog signal by a duty/analog converter 42 and outputted to an output @43. The timer/counter 41 and the duty/analog converter 42 constitute a digital/analog converter 44.

FIG. 2 is a block diagram showing the configuration of the capacitance/time converter in FIG. 1.

A stray capacitance C5 is formed between the moving S pole 7 and the common potential point COM, and the moving amount '#717 is connected to the input end of the inverter G. The output end of inverter G1 is connected to the input end CL of counter CT via inverter G2. One end of the input of Nant gate G3 is counter C
One end of the input of the Nandt gate G4 is connected to the n-bit output @QTL of the counter CT via an inverter G5. Nantes Gate Gco, G4 input fl! ! @ are respectively connected to the output terminals of the inverter G, and these output terminals are respectively connected to the fixed electrodes 9.8. Also, Nantes Gate G3, G
Each input terminal of a Nandt gate G6 is connected between the output terminals of the inverter G6 and the output terminal thereof is connected to the input terminal of the inverter G1 via a bidirectional constant current circuit CC. Note that each element is made of CMO3, for example.

The period T during which the output of the counter CT @QTL is held negative (low level) is the Nant gate G4 and the capacitance C1.
, an inverter G, and a positive feedback loop is formed. Therefore, when the potential at the output terminal of the Nant gate G,1 is E (high level) in the positive power supply voltage, the capacitance C
7 is charged and the input voltage of inverter G rises and reaches the threshold VvH, the potential at the output terminal of inverter G falls vertically and becomes the negative power supply voltage -E (0-level). Therefore, the Nant gate G4 The output terminal of the Nant gate G3 becomes a high level, and the output terminal of the Nant gate G3 remains at a high level as long as the output terminal QTL of the counter CT is held at a low level, so the output terminal of the Nant gate G6 becomes a low level and both A constant current value i from the constant current circuit CC
The charge in the capacitance C4 starts to be discharged. Therefore, the potential at the input end of inverter G1 decreases at a constant rate, and when it reaches the threshold vyH, the potential at the output end of inverter G1 is inverted to a high level. The time t1 from the fall of the input voltage of the inverter G1 until reaching the threshold VTH due to discharge of the bidirectional constant current circuit CC is the capacitance C
Proportional to 1.

Next, when the potential at the output terminal of the inverter G becomes high level, the potential at the output terminal of the Nandt gate G4 becomes low level and the capacitance C1 is charged in the opposite direction. In this case,
The potential at the output terminal of Nant gate G6 becomes high level,
The capacitance C1 is charged by the bidirectional constant current circuit CC with a constant current value i, and the potential at the input end of the inverter G increases at a constant rate, and when it reaches the threshold VTH, the potential at the output end of the inverter G increases. is inverted to low level. The time t until the inverter G1 is inverted is also proportional to the capacitance C1. This operation of 0 or more is repeated until the 8th position of the output terminal QTL of the counter CT becomes a low level. Therefore, the output of the counter CT A period T during which the potential of the r4AQ hole is maintained at a low level.

is a value proportional to the capacitance C1.

When counter CT counts n bits, its output terminal QTL
The potential of is inverted to high level, a positive feedback loop consisting of NAND gate G3, capacitance C2, and inverter G is selected, and oscillation occurs in the same way as when the output terminal of counter CT is held at low level. do. Therefore, counter C
A period T2 during which the output terminal Q of T is held at a high level
is proportional to capacitance C2.

As described above, the capacitance/time converter 31 periodically converts the capacitances C, C2 into times T7 and T2. The converted times T + and T 2 are converted into digital values by the timer counter 32 .

Next, the processing of these converted digital values in FIG. 1 will be explained using the flowchart shown in FIG. 3.

First, the relationship between the capacitances CI and C2 and the pressure P will be explained.

The relative dielectric constant of the sealing liquid sealed between the moving electrode 7 and the fixed pole 8.9 is ε, A is the electrode area of each electrode, and d is the pressure P.
If the distance between each electrode when Δχ is zero is the displacement of the moving electrode 7 due to the pressure P, then the capacitance CI, C2 is C, = εA/(d-Δχ) ... (01) C2= ε
A/(d+Δχ) ...(02), where,
If α is a fixed constant, there is a relationship Δχ = αP, so if we set g = εA/α and h = d/α, equations (01) and (02) become C, = g / (
h − P ) ... (03) C2 = g /
(h + P) ...(04).

Therefore, from these formulas, (CI C2) / (CI + C2) -P/ h
...(05) is obtained. In other words, if the microcomputer unit 30 calculates (CI C2)/(CI + C2), a capacity output V01 proportional to the pressure P can be obtained.

However, in reality, as shown by the broken line in Figure 2, there is a stray capacitance Cs between the moving amount [i7 and the fixed t[i8.9] that does not change at all even if the pressure P that is the measurement input changes. do. In this case, for simplicity, the stray capacitance Cs is the amount of movement'
&7 and the fixed electrode 8.9.

The capacitance output V02 detected in such a case is Vo 2 =K [(CI +C'; ) (C2
+Cs ) ]/[<CI +Cs )+ (C2+
C8) KoK (CI C2)/ (CI +C2+2
C9)...(06) becomes. Substituting equations (03) and (04) into this equation and rearranging it, Vo 2 = (KgP/Cs)/[(gh/Cs
)+h 2-p'> Ko
... (07). Here, if each coefficient is set as a = K g / C6 b = (gh/Cs) + h2, equation (07) becomes Vo 2 = aP/ (b P' ) - (08
). However, this equation (08) shows that the capacitance C1 and 0
2 is ideally made, and in reality, the dimensions do not match, so even if the pressure P is zero, the capacity output V
03 to 0, and considering the coefficient C indicating the amount of offset, equation (08) becomes Vo 3 =aP/ (b P2)
+c-<09).

Since there are three unknowns in equation (09), coefficients a-b and c, it can be determined by calibrating using the three points P and Vc3.

Therefore, next, a calibration procedure for obtaining coefficients a, b, and c will be explained. If the capacitance output V03 when the pressure P is applied as the calibration pressure Pc is vc, three points within the measurement pressure range [Vc (0), Pc (0)] [Vc (1
), Pc (1)], [Vc(2), Pc(2)] into the formula <o9), Vc(○)=aPc(0)/(b-Pc(0) )2
)+C...(10) Vc (1)=aPc (1)/(b-Pc
(1)2)+C...(11) Vc (2)=aPc (2)/(b-Pc
(2) 2 )+C...(12) Obtain. - Numerically, since the input uses Pc = O, here Pc (0) - 0, then (1o) formula % formula % (
13), and by substituting this into equation (11), a=[(Vc
(1)-Vc (0))(b Pc (1)2
)]/Pc (1)...(14) are obtained. Substituting this equation (14) into equation (12), b=Pc (1) ・Pc (2) [(V
c (2) −Vc (0))Pc (2)
(Vc (1)Vc (0)) Pc
(1) ]/[(Vc(2)-Vc(0))Pc
(1) (Vc(1)Vc(0))Pc
(2) Co... (15) By substituting the coefficient s obtained by this equation (15) into the equation (14), the coefficient a is obtained.

The coefficients a, b, and c can be obtained by the coefficient calculation procedure shown in equations (10) to (15) above. This coefficient calculation is performed using the microcomputer unit 30.

In the ROM 36, for example, initial values aO1bo, c of coefficients a, b, and c are stored as initial data in step (3).

etc. are given, and furthermore, in step (2), the calculation procedure of equation (09) or the correction calculation procedure to be described later is stored.

In the above state, the first calibration pressure Pc (0) is applied to the differential capacitance sensor 29, and the capacitances C7 and C2 are read from the differential capacitance sensor 29 under the control of the CPU 37 (step ■). The zero is converted into a signal and stored in the RAM 35 (step ■). Next, in step ■, it is determined whether the number of times the capacitors C1 and C2 have been read has reached three times, and if it has not reached three times, the difference is determined. The second dynamic capacitance sensor 29
After applying the calibration pressure pc(t) for the first time, the process returns to step (2) and the capacitances C1 and C2 are read, and the CPU 37 stores the data in the RAM 35 in the same manner as the first time. Data is measured in the same manner for the third time.

Using the data stored in the RAM 35 in step (2) as described above, the calculations shown in (13) to (15) are executed by the calculation program stored in the ROM 36 in step (2), which is unique to the differential capacitance sensor 29. coefficients a, b, c
is determined and stored in a predetermined area of the RAM 35 (step 2).

Next, using the determined coefficients a, b, and c, a correction calculation formula to be described next is determined.

In order to find the applied measurement pressure PM from the capacitance signal ■- obtained when the measurement pressure is input, P=P
By setting V03=VM and solving for P in equation (09), we get (VM -c )PM 2-aPM+ (VM-c)b
=0, since this is a quadratic equation for PM,
Solving for PM, PH=[a± I (a2 +4 (VM-c)'
b) 112 pieces/-2 (V gate-C) ・
...(16), however, when V11=c, PM=O
It is.

The sign of earth is selected depending on the relationship between VM and prq, but in general, when PM)0 becomes VM)0, - may be selected.

The coefficient data a, b, stored in the RAM 35 in step ①
The correction calculation formula (16) in which c is used is fixedly stored in the ROM 36 functioning as an EEPROM in step (2).

Next, the process moves to step (2), and the CPU 37 calculates the applied measurement pressure P using the correction calculation formula (15) from the capacitance signal VM obtained via the analog/digital converter 33 when the measurement pressure PM is applied. It is calculated, moves to step [phase], and outputs via the digital/analog converter 44. Thereafter, repeat steps ① and ② to measure the pressure.

The relationship between the output value and input value output after such correction calculations is extremely good, as can be seen from the actual characteristics shown by the broken line and the correction curve shown by the dotted line, as shown in Figure 4 (a). The output error is extremely small as shown in FIG. 4 (b).

Although the case of a pressure transmitter has been described above, the present invention can be similarly applied to a differential pressure transmitter. However, in the case of a differential transmitter, the measured pressure spans both the positive and negative sides, and the movement of the moving electrode is different between the positive and negative sides and is not completely symmetrical. Therefore, in the calibration procedure for calculating coefficients a, b=c, in addition to the coefficients a-b, c found at the three points on the positive input side, the coefficient a\b\C- found on the negative input side is required. However, since the zero input point can be shared, C=C-, and calibration is performed using five points. In other words, at five calibration points, coefficients a, b, C, a-
1b-, and when Vc>c, use coefficients a, b, and c to find Pc, and when V(=c, Pc=O, and V
When C<C, the coefficient a-1b-1c is used to find Pc, and even with a differential pressure transmitter, the fifth
As shown in the figure, accuracy can be improved.

<Effects of the Invention> As explained above in detail with the embodiments, the coefficients of the correction curve are determined by the coefficient calculation means by applying the calibration input to the predetermined calculation formula, and the correction calculation means uses the coefficients to calculate the differential capacitance. Since the correction curve for correcting the characteristics of the sensor is specified and calculated, highly accurate correction can be performed with a small number of calibration points, which improves work efficiency and contributes to cost reduction.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of one embodiment of the present invention, FIG. 2 is a circuit diagram showing a specific configuration of the capacitance/time converter in FIG. 1, and FIG. 3 is similar to FIG. 1. FIG. 4 is a characteristic diagram explaining the effect of the embodiment shown in FIG. 1, and FIG. 5 is a flowchart explaining the operation of the embodiment shown in FIG. Characteristic diagram, 6th
The figure is a characteristic diagram showing the characteristics in the case of conventional polygonal line approximation.
The figure is a characteristic diagram showing the characteristics in the case of conventional polynomial approximation. 7... Moving electrode, 8.9... Fixed electrode, 29...
Differential capacitance sensor, 30... Microprocessor unit, 33... Analog/digital converter, 35...
Random access memory, 36... Read only memory, 37... Processor, 44... Digital/analog converter. Figure 31 Capacitance/B Now PiI change required 7' Ward σ) Figure 4

Claims (1)

[Claims] A first electrode and a second electrode are provided to face the movable electrode, and a sealing liquid is filled between these electrodes, and the second electrode changes differentially in accordance with the pressure to be detected. a differential capacitance sensor forming first and second capacitances; a microcomputer unit that converts capacitance signals corresponding to the first and second capacitances into digital signals and performs arithmetic processing; In a pressure transmitter that converts the calculation result into an analog signal and outputs it, the microcomputer unit generates three capacitance signals Vc obtained for known calibration pressures Pc at arbitrary three points within the pressure measurement range. is digitally converted and stored in the memory, and using these stored data pairs (Pc, Vc), determine the coefficients a, b, and c of the formula Vc = [aPc/(b-Pc^2)] + c. The formula P_M=[a±{( a^2+4(V_M-c)^2b}
^1^/^2]/-2(V_M-c), (however, when V_M=c, P_M=O), and a correction calculation means for outputting the measured pressure P_M whose input/output characteristics have been corrected by the following. A pressure transmitter characterized by: